Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This st...Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This study proposes a novel end-to-end disparity estimation model to address these challenges.Our approach combines a Pseudo-Siamese neural network architecture with pyramid dilated convolutions,integrating multi-scale image information to enhance robustness against lighting interferences.This study introduces a Pseudo-Siamese structure-based disparity regression model that simplifies left-right image comparison,improving accuracy and efficiency.The model was evaluated using a dataset of stereo endoscopic videos captured by the Da Vinci surgical robot,comprising simulated silicone heart sequences and real heart video data.Experimental results demonstrate significant improvement in the network’s resistance to lighting interference without substantially increasing parameters.Moreover,the model exhibited faster convergence during training,contributing to overall performance enhancement.This study advances endoscopic image processing accuracy and has potential implications for surgical robot applications in complex environments.展开更多
软测量技术为工业过程中重要变量及难测变量的预测提供了一个有效的解决办法。然而,由于工业过程的复杂化和高昂的数据获取成本,使得标记数据与未标记数据分布不平衡。此时,构建高性能的软测量模型成为一个挑战。针对这一问题,提出了一...软测量技术为工业过程中重要变量及难测变量的预测提供了一个有效的解决办法。然而,由于工业过程的复杂化和高昂的数据获取成本,使得标记数据与未标记数据分布不平衡。此时,构建高性能的软测量模型成为一个挑战。针对这一问题,提出了一种基于时差的多输出tri-training异构软测量方法。通过构建一种新的tri-training框架,采用多输出的高斯过程回归(multi-output Gaussian process regression,MGPR)、相关向量机(multi-output relevance vector machine,MRVM)、最小二乘支持向量机(multi-output least squares support vector machine,MLSSVM)三种模型作为基线监督回归器,使用标记数据进行训练和迭代;同时,引入时间差分(time difference,TD)改进模型的动态特性,并通过卡尔曼滤波(Kalman filtering,KF)优化模型的参数,提高其预测性能;最后通过模拟污水处理平台(benchmark simulation model 1,BSM1)和实际污水处理厂对该模型进行了验证。结果表明,与传统的软测量建模方法相比,该模型能显著提高数据分布不平衡下软测量模型的自适应性和预测性能。展开更多
BACKGROUND Cognitive frailty and depression are prevalent among the elderly,significantly impairing physical and cognitive functions,psychological well-being,and quality of life.Effective interventions are essential t...BACKGROUND Cognitive frailty and depression are prevalent among the elderly,significantly impairing physical and cognitive functions,psychological well-being,and quality of life.Effective interventions are essential to mitigate these adverse effects and enhance overall health outcomes in this population.AIM To evaluate the effects of exercise-cognitive dual-task training on frailty,cognitive function,psychological status,and quality of life in elderly patients with cognitive frailty and depression.METHODS A retrospective study was conducted on 130 patients with cognitive frailty and depression admitted between December 2021 and December 2023.Patients were divided into a control group receiving routine intervention and an observation group undergoing exercise-cognitive dual-task training in addition to routine care.Frailty,cognitive function,balance and gait,psychological status,and quality of life were assessed before and after the intervention.RESULTS After the intervention,the frailty score of the observation group was(5.32±0.69),lower than that of the control group(5.71±0.55).The Montreal cognitive assessment basic scale score in the observation group was(24.06±0.99),higher than the control group(23.43±1.40).The performance oriented mobility assessment score in the observation group was(21.81±1.24),higher than the control group(21.15±1.26).The self-efficacy in the observation group was(28.27±2.66),higher than the control group(30.05±2.66).The anxiety score in the hospital anxiety and depression scale(HADS)for the observation group was(5.86±0.68),lower than the control group(6.21±0.64).The depression score in the HADS for the observation group was(5.67±0.75),lower than the control group(6.27±0.92).Additionally,the scores for each dimension of the 36-item short form survey in the observation group were higher than those in the control group,with statistically significant differences(P<0.05).CONCLUSION Exercise-cognitive dual-task training is beneficial for improving frailty,enhancing cognitive function,and improving psychological status and quality of life in elderly patients with cognitive frailty and depression.展开更多
Exercise is a therapeutic approach in cancer treatment,providing several benefits.Moreover,exercise is associated with a reduced risk for developing a range of cancers and for their recurrence,as well as with improvin...Exercise is a therapeutic approach in cancer treatment,providing several benefits.Moreover,exercise is associated with a reduced risk for developing a range of cancers and for their recurrence,as well as with improving survival,even though the underlying mechanisms remain unclear.Preclinical and clinical evidence shows that the acute effects of a single exercise session can suppress the growth of various cancer cell lines in vitro.This suppression is potentially due to altered concentrations of hormones(e.g.,insulin)and cytokines(e.g.,tumor necrosis factor alpha and interleukin 6)after exercise.These factors,known to be involved in tumorigenesis,may explain why exercise is associated with reduced cancer incidence,recurrence,and mortality.However,the effects of short-(<8 weeks)and long-term(≥8 weeks)exercise programs on cancer cells have been reported with mixed results.Although more research is needed,it appears that interventions incorporating both exercise and diet seem to have greater inhibitory effects on cancer cell growth in both apparently healthy subjects as well as in cancer patients.Although speculative,these suppressive effects on cancer cells may be driven by changes in body weight and composition as well as by a reduction in low-grade inflammation often associated with sedentary behavior,low muscle mass,and excess fat mass in cancer patients.Taken together,such interventions could alter the systemic levels of suppressive circulating factors,leading to a less favorable environment for tumorigenesis.While regular exercise and a healthy diet may establish a more cancer-suppressive environment,each acute bout of exercise provides a further“dose”of anticancer medicine.Therefore,integrating regular exercise could potentially play a significant role in cancer management,highlighting the need for future investigations in this promising area of research.展开更多
Seismic data denoising is a critical process usually applied at various stages of the seismic processing workflow,as our ability to mitigate noise in seismic data affects the quality of our subsequent analyses.However...Seismic data denoising is a critical process usually applied at various stages of the seismic processing workflow,as our ability to mitigate noise in seismic data affects the quality of our subsequent analyses.However,finding an optimal balance between preserving seismic signals and effectively reducing seismic noise presents a substantial challenge.In this study,we introduce a multi-stage deep learning model,trained in a self-supervised manner,designed specifically to suppress seismic noise while minimizing signal leakage.This model operates as a patch-based approach,extracting overlapping patches from the noisy data and converting them into 1D vectors for input.It consists of two identical sub-networks,each configured differently.Inspired by the transformer architecture,each sub-network features an embedded block that comprises two fully connected layers,which are utilized for feature extraction from the input patches.After reshaping,a multi-head attention module enhances the model’s focus on significant features by assigning higher attention weights to them.The key difference between the two sub-networks lies in the number of neurons within their fully connected layers.The first sub-network serves as a strong denoiser with a small number of neurons,effectively attenuating seismic noise;in contrast,the second sub-network functions as a signal-add-back model,using a larger number of neurons to retrieve some of the signal that was not preserved in the output of the first sub-network.The proposed model produces two outputs,each corresponding to one of the sub-networks,and both sub-networks are optimized simultaneously using the noisy data as the label for both outputs.Evaluations conducted on both synthetic and field data demonstrate the model’s effectiveness in suppressing seismic noise with minimal signal leakage,outperforming some benchmark methods.展开更多
Computed Tomography(CT)reconstruction is essential inmedical imaging and other engineering fields.However,blurring of the projection during CT imaging can lead to artifacts in the reconstructed images.Projection blur ...Computed Tomography(CT)reconstruction is essential inmedical imaging and other engineering fields.However,blurring of the projection during CT imaging can lead to artifacts in the reconstructed images.Projection blur combines factors such as larger ray sources,scattering and imaging system vibration.To address the problem,we propose DeblurTomo,a novel self-supervised learning-based deblurring and reconstruction algorithm that efficiently reconstructs sharp CT images from blurry input without needing external data and blur measurement.Specifically,we constructed a coordinate-based implicit neural representation reconstruction network,which can map the coordinates to the attenuation coefficient in the reconstructed space formore convenient ray representation.Then,wemodel the blur as aweighted sumof offset rays and design the RayCorrectionNetwork(RCN)andWeight ProposalNetwork(WPN)to fit these rays and their weights bymulti-view consistency and geometric information,thereby extending 2D deblurring to 3D space.In the training phase,we use the blurry input as the supervision signal to optimize the reconstruction network,the RCN,and the WPN simultaneously.Extensive experiments on the widely used synthetic dataset show that DeblurTomo performs superiorly on the limited-angle and sparse-view in the simulated blurred scenarios.Further experiments on real datasets demonstrate the superiority of our method in practical scenarios.展开更多
Standardized residency training programs primarily focus on developing clinical diagnostic and treatment skills,often allocating limited time to research activities.However,enhancing research skills is of paramount im...Standardized residency training programs primarily focus on developing clinical diagnostic and treatment skills,often allocating limited time to research activities.However,enhancing research skills is of paramount importance for residents,as it fosters critical thinking,problem-solving abilities,and a deeper understanding of applying scientific principles to clinical practice.This paper explores the necessity and significance of integrating research training into residency programs,emphasizing its role in cultivating well-rounded physicians capable of advancing medical knowledge.This study proposes a competency-based research training model that encompasses research literacy,study design,biostatistics,and scientific writing.Additionally,online asynchronous training modules,robust mentorship,and balanced time management strategies are recommended to enhance residents’research engagement without compromising clinical training.By implementing these measures,residency programs can improve residents’research capabilities,contributing to both individual professional growth and the broader advancement of medical science.展开更多
Purpose We aimed to determine:(a)the chronic effects of interval training(IT)combined with blood flow restriction(BFR)on physiological adaptations(aerobic/anaerobic capacity and muscle responses)and performance enhanc...Purpose We aimed to determine:(a)the chronic effects of interval training(IT)combined with blood flow restriction(BFR)on physiological adaptations(aerobic/anaerobic capacity and muscle responses)and performance enhancement(endurance and sprints),and(b)the influence of participant characteristics and intervention protocols on these effects.Methods Searches were conducted in PubMed,Web of Science(Core Collection),Cochrane Library(Embase,ClinicalTrials.gov,and International Clinical Trials Registry Platform),and Chinese National Knowledge Infrastructure on April 2,with updates on October 17,2024.Pooled effects for each outcome were summarized using Hedge's g(g)through meta-analysis-based random effects models,and subgroup and regression analyses were used to explore moderators.Results A total of 24 studies with 621 participants were included.IT combined with BFR(IT+BFR)significantly improved maximal oxygen uptake(VO2_(max))(g=0.63,I^(2)=63%),mean power during the Wingate 30-s test(g=0.70,I^(2)=47%),muscle strength(g=0.88,I^(2)=64%),muscle endurance(g=0.43,I^(2)=0%),time to fatigue(g=1.26,I^(2)=86%),and maximal aerobic speed(g=0.74,I^(2)=0%)compared to IT alone.Subgroup analysis indicated that participant characteristics including training status,IT intensity,and IT modes significantly moderated VO2_(max)(subgroup differences:p<0.05).Specifically,IT+BFR showed significantly superior improvements in VO2_(max)compared to IT alone in trained individuals(g=0.76)at supra-maximal intensity(g=1.29)and moderate intensity(g=1.08)as well as in walking(g=1.64)and running(g=0.63)modes.Meta-regression analysis showed cuff width(β=0.14)was significantly associated with VO2_(max)change,identifying 8.23 cm as the minimum threshold required for significant improvement.Subgroup analyses regarding muscle strength did not reveal any significant moderators.Conclusion IT+BFR enhances physiological adaptations and optimizes aspects of endurance performance,with moderators including training status,IT protocol(intensity,mode,and type),and cuff width.This intervention addresses various IT-related challenges and provides tailored protocols and benefits for diverse populations.展开更多
This paper reports a case of cerebral stem infarction with quadriplegia and complete dependence on daily life.The course of the disease lasted more than 7 months.Frenchay's improved articulation Disorder Assessmen...This paper reports a case of cerebral stem infarction with quadriplegia and complete dependence on daily life.The course of the disease lasted more than 7 months.Frenchay's improved articulation Disorder Assessment Form has been assessed as severe articulation disorder.The patient has significantly improved his speech function and quality of life after systematic head control training,respiratory function training,articulation motor training,and articulation training.In the course of treatment,emphasis was placed on head postural control training and respiratory function training,and emphasis was placed on the strength and coordination training of articulatory organs,and the results were remarkable.After the patient was discharged from the hospital,the follow-up of basic daily life communication was not limited.展开更多
Blended acquisition offers efficiency improvements over conventional seismic data acquisition, at the cost of introducing blending noise effects. Besides, seismic data often suffers from irregularly missing shots caus...Blended acquisition offers efficiency improvements over conventional seismic data acquisition, at the cost of introducing blending noise effects. Besides, seismic data often suffers from irregularly missing shots caused by artificial or natural effects during blended acquisition. Therefore, blending noise attenuation and missing shots reconstruction are essential for providing high-quality seismic data for further seismic processing and interpretation. The iterative shrinkage thresholding algorithm can help obtain deblended data based on sparsity assumptions of complete unblended data, and it characterizes seismic data linearly. Supervised learning algorithms can effectively capture the nonlinear relationship between incomplete pseudo-deblended data and complete unblended data. However, the dependence on complete unblended labels limits their practicality in field applications. Consequently, a self-supervised algorithm is presented for simultaneous deblending and interpolation of incomplete blended data, which minimizes the difference between simulated and observed incomplete pseudo-deblended data. The used blind-trace U-Net (BTU-Net) prevents identity mapping during complete unblended data estimation. Furthermore, a multistep process with blending noise simulation-subtraction and missing traces reconstruction-insertion is used in each step to improve the deblending and interpolation performance. Experiments with synthetic and field incomplete blended data demonstrate the effectiveness of the multistep self-supervised BTU-Net algorithm.展开更多
Background Schizophrenia is characterised by pervasive cognitive deficits that significantly impair daily functioning and quality of life.Pharmacological treatments have limited efficacy in addressing these deficits,h...Background Schizophrenia is characterised by pervasive cognitive deficits that significantly impair daily functioning and quality of life.Pharmacological treatments have limited efficacy in addressing these deficits,highlighting the need for adjunctive interventions like computerised cognitive training(CCT).Aims This study aimed to evaluate the effects of a 30-session CCT programme on mental well-being and cognitive performance in individuals with schizophrenia.Additionally,it assessed the usability and acceptability of CCT in this population.Methods A double-blind,randomised clinical trial was conducted with 54 participants assigned to intervention and control groups.Cognitive and mental health outcomes were assessed using validated tools such as the Depression Anxiety Stress Scale 21,the Warwick-Edinburgh Mental Wellbeing Scale and the Cambridge Neuropsychological Test Automated Battery.Usability was measured with the System Usability Scale(SUS).Assessments were conducted at baseline,post-intervention and 3 months post-follow-up.Results The CCT intervention significantly improved mental well-being,reduced stress and enhanced working memory(paired associate learning,spatial working memory and spatial span)compared with controls.However,no significant effects were observed for anxiety,depression or executive function.Usability scores were high(SUS=83.51),and compliance rates were strong(92.7%),indicating favourable participant engagement.Conclusion CCT demonstrated potential as an adjunctive treatment for schizophrenia,with significant improvements in targeted cognitive and mental health domains.The high usability and compliance rates support its feasibility for broader implementation.Further research is needed to optimise protocols and explore long-term benefits.CCT offers a promising approach to addressing mental health and cognitive challenges in schizophrenia,particularly for stress and working memory.Its usability and acceptability suggest it could be seamlessly integrated into clinical practice.展开更多
In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operati...In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.展开更多
Objective:The objective of this study is to develop and evaluate a structured clinical rotation-based emergency response training program to enhance nurses’emergency competence,theoretical knowledge,and technical ski...Objective:The objective of this study is to develop and evaluate a structured clinical rotation-based emergency response training program to enhance nurses’emergency competence,theoretical knowledge,and technical skills.Methods:A comprehensive emergency training program was developed,and a randomized controlled trial was implemented from June 2022 to May 2023 at a tertiary general hospital in Chongqing,China.The study involved 214 nurses,with 106 participants in the intervention group receiving a 3-month innovative emergency response competence training and 108 in the control group undergoing conventional training.Postintervention assessments evaluated emergency response capabilities using the Emergency Response Ability Assessment Scale for Nurses in Public Health Emergencies,theoretical knowledge through a self-designed comprehensive theoretical assessment instrument,technical skills using a standardized skill assessment form,and training satisfaction through two distinct feedback instruments.Results:The emergency response ability scores were significantly higher in the intervention group compared to controls(3.99±0.18 vs.2.53±0.25,P<0.05).Theoretical assessment scores showed marked improvement in the intervention group versus the control group(85.31±4.40 vs.52.45±6.19,P<0.05).Technical skill performance was significantly better in the intervention group than that in controls(94.47±1.64 vs.86.39±2.36,P<0.05).Training satisfaction was higher among intervention group nurses compared to controls(4.53±0.23 vs.4.00±0.38,P<0.05),with nursing managers also reporting greater satisfaction with the intervention program versus conventional training(4.57±0.49 vs.3.92±0.79,P<0.05).Conclusion:The clinical rotation-based structured emergency response training program effectively enhances nurses’emergency competencies,theoretical knowledge,and technical skills.These findings provide both theoretical foundations and practical guidelines for developing emergency response and specialized nursing competence training programs.展开更多
Endoscopic ultrasound(EUS)is an indispensable tool for the diagnosis and management of various diseases,particularly biliopancreatic disorders,as it provides detailed visualization of the gastrointestinal tract and su...Endoscopic ultrasound(EUS)is an indispensable tool for the diagnosis and management of various diseases,particularly biliopancreatic disorders,as it provides detailed visualization of the gastrointestinal tract and surrounding structures.As the demand for diagnostic and interventional EUS procedures increases,ensuring high-quality training for endoscopists is essential to improve patient outcomes.This mini-review provides an overview of the current state of EUS training and emphasizes the importance of a structured approach that integrates theoretical knowledge and hands-on experience.We discuss different training methods,focusing on the main courses available worldwide,and highlight their advantages and limitations.In addition,we examine the challenges of training for diagnostic and interventional EUS,such as limited access to training centers and the need for personalized feedback.Overall,improving EUS training programs is essential to enhance physician skills and ensure this advanced technique is used safely and efficiently in clinical practice.展开更多
Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely us...Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely used in pulmonary disease diagnosis,such as pneumonia and tuberculosis.However,traditional feature fusion methods often suffer from feature disparity,information loss,redundancy,and increased complexity,hindering the further extension of DL algorithms.To solve this problem,we propose a Graph-Convolution Fusion Network with Self-Supervised Feature Alignment(Self-FAGCFN)to address the limitations of traditional feature fusion methods in deep learning-based medical image classification for respiratory diseases such as pneumonia and tuberculosis.The network integrates Convolutional Neural Networks(CNNs)for robust feature extraction from two-dimensional grid structures and Graph Convolutional Networks(GCNs)within a Graph Neural Network branch to capture features based on graph structure,focusing on significant node representations.Additionally,an Attention-Embedding Ensemble Block is included to capture critical features from GCN outputs.To ensure effective feature alignment between pre-and post-fusion stages,we introduce a feature alignment loss that minimizes disparities.Moreover,to address the limitations of proposed methods,such as inappropriate centroid discrepancies during feature alignment and class imbalance in the dataset,we develop a Feature-Centroid Fusion(FCF)strategy and a Multi-Level Feature-Centroid Update(MLFCU)algorithm,respectively.Extensive experiments on public datasets LungVision and Chest-Xray demonstrate that the Self-FAGCFN model significantly outperforms existing methods in diagnosing pneumonia and tuberculosis,highlighting its potential for practical medical applications.展开更多
Here we compare the efficacy of anti-obesity drugs alone or combined with exercise training on body weight and exercise capacity of obese patients.Randomized clinical trials that assessed the impact of any anti-obesit...Here we compare the efficacy of anti-obesity drugs alone or combined with exercise training on body weight and exercise capacity of obese patients.Randomized clinical trials that assessed the impact of any anti-obesity drug alone or combined with exercise training on body weight,body fat,fat-free mass and cardiorespiratory fitness in obese patients were retrieved from Pubmed and EMBASE up to May 2024.Risk of bias assessment was performed with RoB 2.0,and the GRADE approach assessed the certainty of evidence(CoE)of each main outcome.We included four publications summing up 202 patients.Two publications used orlistat as an anti-obesity drug treatment,while the other two adopted GLP-1 receptor agonist(liraglutide or tirzepatide)as a pharmacotherapy for weight management.Orlistat combined with exercise was superior to change body weight(mean difference(MD):−2.27 kg;95%CI:−2.86 to−1.69;CoE:very low),fat mass(MD:−2.89;95%CI:−3.87 to−1.91;CoE:very low),fat-free mass(MD:0.56;95%CI:0.40–0.72;CoE:very low),and VO_(2)Peak(MD:2.64;95%CI:2.52–2.76;CoE:very low).GLP-1 receptor agonist drugs combined with exercise had a great effect on body weight(MD:−3.96 kg;95%CI:−5.07 to−2.85;CoE:low),fat mass(MD:−1.76;95%CI:−2.24 to−1.27;CoE:low),fat-free mass(MD:0.50;95%CI:−0.98 to 1.98;CoE:very low)and VO_(2)Peak(MD:2.47;95%CI:1.31–3.63;CoE:very low).The results reported here suggest that exercise training remains an important approach in weight management when combined with pharmacological treatment.展开更多
Self-supervised monocular depth estimation has emerged as a major research focus in recent years,primarily due to the elimination of ground-truth depth dependence.However,the prevailing architectures in this domain su...Self-supervised monocular depth estimation has emerged as a major research focus in recent years,primarily due to the elimination of ground-truth depth dependence.However,the prevailing architectures in this domain suffer from inherent limitations:existing pose network branches infer camera ego-motion exclusively under static-scene and Lambertian-surface assumptions.These assumptions are often violated in real-world scenarios due to dynamic objects,non-Lambertian reflectance,and unstructured background elements,leading to pervasive artifacts such as depth discontinuities(“holes”),structural collapse,and ambiguous reconstruction.To address these challenges,we propose a novel framework that integrates scene dynamic pose estimation into the conventional self-supervised depth network,enhancing its ability to model complex scene dynamics.Our contributions are threefold:(1)a pixel-wise dynamic pose estimation module that jointly resolves the pose transformations of moving objects and localized scene perturbations;(2)a physically-informed loss function that couples dynamic pose and depth predictions,designed to mitigate depth errors arising from high-speed distant objects and geometrically inconsistent motion profiles;(3)an efficient SE(3)transformation parameterization that streamlines network complexity and temporal pre-processing.Extensive experiments on the KITTI and NYU-V2 benchmarks show that our framework achieves state-of-the-art performance in both quantitative metrics and qualitative visual fidelity,significantly improving the robustness and generalization of monocular depth estimation under dynamic conditions.展开更多
Digital economy has become a new driving force for China’s economic growth,continuously boosting economic development and rapidly integrating into various fields of China’s economy and society.The advent of the digi...Digital economy has become a new driving force for China’s economic growth,continuously boosting economic development and rapidly integrating into various fields of China’s economy and society.The advent of the digital economy era has reshaped the development pattern of the financial industry.The rapid development of financial technology has promoted the transformation of financial formats and put forward higher requirements for financial talent training in the new era.Digital finance is not only a key part of the transformation and upgrading of China’s financial industry but also an integral component of China’s modern financial ecosystem.Against the backdrop of the digital economy,cultivating digital finance talents in vocational colleges is of great significance to China’s economic development.This paper analyzes the predicaments faced in digital finance talent training,explores in depth the reform of digital finance talent training modes,and proposes policy suggestions for the digital finance talent training system based on the development characteristics of digital finance.展开更多
High dropout rates in short-term job skills training programs hinder workforce development.This study applies machine learning to predict program completion while addressing class imbalance challenges.A dataset of6548...High dropout rates in short-term job skills training programs hinder workforce development.This study applies machine learning to predict program completion while addressing class imbalance challenges.A dataset of6548 records with 24 demographic,educational,program-specific,and employment-related features was analyzed.Data preprocessing involved cleaning,encoding categorical variables,and balancing the dataset using the Synthetic Minority Oversampling Technique(SMOTE),as only 15.9% of participants were dropouts.six machine learning models-Logistic Regression,Random Forest,SupportVector Machine,K-Nearest Neighbors,Naive Bayes,and XGBoost-were evaluated on both balanced and unbalanced datasets using an 80-20 train-test split.Performance was assessed using Accuracy,Precision,Recall,F1-score,and ROC-AUC.XGBoost achieved the highest performance on the balanced dataset,with an F1-score of 0.9200 and aROC-AUC of0.9684,followed by Random Forest.These findings highlight the potential of machine learning for early identification of dropout trainees,aiding in retention strategies for workforce training.The results support the integration of predictive analytics to optimize intervention efforts in short-term training programs.展开更多
基金Supported by Sichuan Science and Technology Program(2023YFSY0026,2023YFH0004)Supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korean government(MSIT)(No.RS-2022-00155885,Artificial Intelligence Convergence Innovation Human Resources Development(Hanyang University ERICA)).
文摘Two-dimensional endoscopic images are susceptible to interferences such as specular reflections and monotonous texture illumination,hindering accurate three-dimensional lesion reconstruction by surgical robots.This study proposes a novel end-to-end disparity estimation model to address these challenges.Our approach combines a Pseudo-Siamese neural network architecture with pyramid dilated convolutions,integrating multi-scale image information to enhance robustness against lighting interferences.This study introduces a Pseudo-Siamese structure-based disparity regression model that simplifies left-right image comparison,improving accuracy and efficiency.The model was evaluated using a dataset of stereo endoscopic videos captured by the Da Vinci surgical robot,comprising simulated silicone heart sequences and real heart video data.Experimental results demonstrate significant improvement in the network’s resistance to lighting interference without substantially increasing parameters.Moreover,the model exhibited faster convergence during training,contributing to overall performance enhancement.This study advances endoscopic image processing accuracy and has potential implications for surgical robot applications in complex environments.
文摘软测量技术为工业过程中重要变量及难测变量的预测提供了一个有效的解决办法。然而,由于工业过程的复杂化和高昂的数据获取成本,使得标记数据与未标记数据分布不平衡。此时,构建高性能的软测量模型成为一个挑战。针对这一问题,提出了一种基于时差的多输出tri-training异构软测量方法。通过构建一种新的tri-training框架,采用多输出的高斯过程回归(multi-output Gaussian process regression,MGPR)、相关向量机(multi-output relevance vector machine,MRVM)、最小二乘支持向量机(multi-output least squares support vector machine,MLSSVM)三种模型作为基线监督回归器,使用标记数据进行训练和迭代;同时,引入时间差分(time difference,TD)改进模型的动态特性,并通过卡尔曼滤波(Kalman filtering,KF)优化模型的参数,提高其预测性能;最后通过模拟污水处理平台(benchmark simulation model 1,BSM1)和实际污水处理厂对该模型进行了验证。结果表明,与传统的软测量建模方法相比,该模型能显著提高数据分布不平衡下软测量模型的自适应性和预测性能。
文摘BACKGROUND Cognitive frailty and depression are prevalent among the elderly,significantly impairing physical and cognitive functions,psychological well-being,and quality of life.Effective interventions are essential to mitigate these adverse effects and enhance overall health outcomes in this population.AIM To evaluate the effects of exercise-cognitive dual-task training on frailty,cognitive function,psychological status,and quality of life in elderly patients with cognitive frailty and depression.METHODS A retrospective study was conducted on 130 patients with cognitive frailty and depression admitted between December 2021 and December 2023.Patients were divided into a control group receiving routine intervention and an observation group undergoing exercise-cognitive dual-task training in addition to routine care.Frailty,cognitive function,balance and gait,psychological status,and quality of life were assessed before and after the intervention.RESULTS After the intervention,the frailty score of the observation group was(5.32±0.69),lower than that of the control group(5.71±0.55).The Montreal cognitive assessment basic scale score in the observation group was(24.06±0.99),higher than the control group(23.43±1.40).The performance oriented mobility assessment score in the observation group was(21.81±1.24),higher than the control group(21.15±1.26).The self-efficacy in the observation group was(28.27±2.66),higher than the control group(30.05±2.66).The anxiety score in the hospital anxiety and depression scale(HADS)for the observation group was(5.86±0.68),lower than the control group(6.21±0.64).The depression score in the HADS for the observation group was(5.67±0.75),lower than the control group(6.27±0.92).Additionally,the scores for each dimension of the 36-item short form survey in the observation group were higher than those in the control group,with statistically significant differences(P<0.05).CONCLUSION Exercise-cognitive dual-task training is beneficial for improving frailty,enhancing cognitive function,and improving psychological status and quality of life in elderly patients with cognitive frailty and depression.
文摘Exercise is a therapeutic approach in cancer treatment,providing several benefits.Moreover,exercise is associated with a reduced risk for developing a range of cancers and for their recurrence,as well as with improving survival,even though the underlying mechanisms remain unclear.Preclinical and clinical evidence shows that the acute effects of a single exercise session can suppress the growth of various cancer cell lines in vitro.This suppression is potentially due to altered concentrations of hormones(e.g.,insulin)and cytokines(e.g.,tumor necrosis factor alpha and interleukin 6)after exercise.These factors,known to be involved in tumorigenesis,may explain why exercise is associated with reduced cancer incidence,recurrence,and mortality.However,the effects of short-(<8 weeks)and long-term(≥8 weeks)exercise programs on cancer cells have been reported with mixed results.Although more research is needed,it appears that interventions incorporating both exercise and diet seem to have greater inhibitory effects on cancer cell growth in both apparently healthy subjects as well as in cancer patients.Although speculative,these suppressive effects on cancer cells may be driven by changes in body weight and composition as well as by a reduction in low-grade inflammation often associated with sedentary behavior,low muscle mass,and excess fat mass in cancer patients.Taken together,such interventions could alter the systemic levels of suppressive circulating factors,leading to a less favorable environment for tumorigenesis.While regular exercise and a healthy diet may establish a more cancer-suppressive environment,each acute bout of exercise provides a further“dose”of anticancer medicine.Therefore,integrating regular exercise could potentially play a significant role in cancer management,highlighting the need for future investigations in this promising area of research.
基金supported by the King Abdullah University of Science and Technology(KAUST)。
文摘Seismic data denoising is a critical process usually applied at various stages of the seismic processing workflow,as our ability to mitigate noise in seismic data affects the quality of our subsequent analyses.However,finding an optimal balance between preserving seismic signals and effectively reducing seismic noise presents a substantial challenge.In this study,we introduce a multi-stage deep learning model,trained in a self-supervised manner,designed specifically to suppress seismic noise while minimizing signal leakage.This model operates as a patch-based approach,extracting overlapping patches from the noisy data and converting them into 1D vectors for input.It consists of two identical sub-networks,each configured differently.Inspired by the transformer architecture,each sub-network features an embedded block that comprises two fully connected layers,which are utilized for feature extraction from the input patches.After reshaping,a multi-head attention module enhances the model’s focus on significant features by assigning higher attention weights to them.The key difference between the two sub-networks lies in the number of neurons within their fully connected layers.The first sub-network serves as a strong denoiser with a small number of neurons,effectively attenuating seismic noise;in contrast,the second sub-network functions as a signal-add-back model,using a larger number of neurons to retrieve some of the signal that was not preserved in the output of the first sub-network.The proposed model produces two outputs,each corresponding to one of the sub-networks,and both sub-networks are optimized simultaneously using the noisy data as the label for both outputs.Evaluations conducted on both synthetic and field data demonstrate the model’s effectiveness in suppressing seismic noise with minimal signal leakage,outperforming some benchmark methods.
基金supported in part by the National Natural Science Foundation of China under Grants 62472434 and 62402171in part by the National Key Research and Development Program of China under Grant 2022YFF1203001+1 种基金in part by the Science and Technology Innovation Program of Hunan Province under Grant 2022RC3061in part by the Sci-Tech Innovation 2030 Agenda under Grant 2023ZD0508600.
文摘Computed Tomography(CT)reconstruction is essential inmedical imaging and other engineering fields.However,blurring of the projection during CT imaging can lead to artifacts in the reconstructed images.Projection blur combines factors such as larger ray sources,scattering and imaging system vibration.To address the problem,we propose DeblurTomo,a novel self-supervised learning-based deblurring and reconstruction algorithm that efficiently reconstructs sharp CT images from blurry input without needing external data and blur measurement.Specifically,we constructed a coordinate-based implicit neural representation reconstruction network,which can map the coordinates to the attenuation coefficient in the reconstructed space formore convenient ray representation.Then,wemodel the blur as aweighted sumof offset rays and design the RayCorrectionNetwork(RCN)andWeight ProposalNetwork(WPN)to fit these rays and their weights bymulti-view consistency and geometric information,thereby extending 2D deblurring to 3D space.In the training phase,we use the blurry input as the supervision signal to optimize the reconstruction network,the RCN,and the WPN simultaneously.Extensive experiments on the widely used synthetic dataset show that DeblurTomo performs superiorly on the limited-angle and sparse-view in the simulated blurred scenarios.Further experiments on real datasets demonstrate the superiority of our method in practical scenarios.
基金Research on the Construction of a Course Ideological and Political Education System and Evaluation Framework for the“Traditional Chinese Medicine Internal Medicine”Course(Project No.:2025J0459)Open Research Fund Program of Yunnan Key Laboratory of Integrated Traditional Chinese(Project No.:CWCD2023-002,CWCD2023-003&CWCD2023-009)+2 种基金Western Medicine for Chronic Disease in Prevention and TreatmentYunnan 047 Key Laboratory of Yi Dai Medicine and Yi Medicine(Project No.:2024SS24047)Yunnan 025 Key Laboratory of Yi Dai Medicine and Yi Medicine(Project No.:2024SS24025)。
文摘Standardized residency training programs primarily focus on developing clinical diagnostic and treatment skills,often allocating limited time to research activities.However,enhancing research skills is of paramount importance for residents,as it fosters critical thinking,problem-solving abilities,and a deeper understanding of applying scientific principles to clinical practice.This paper explores the necessity and significance of integrating research training into residency programs,emphasizing its role in cultivating well-rounded physicians capable of advancing medical knowledge.This study proposes a competency-based research training model that encompasses research literacy,study design,biostatistics,and scientific writing.Additionally,online asynchronous training modules,robust mentorship,and balanced time management strategies are recommended to enhance residents’research engagement without compromising clinical training.By implementing these measures,residency programs can improve residents’research capabilities,contributing to both individual professional growth and the broader advancement of medical science.
文摘Purpose We aimed to determine:(a)the chronic effects of interval training(IT)combined with blood flow restriction(BFR)on physiological adaptations(aerobic/anaerobic capacity and muscle responses)and performance enhancement(endurance and sprints),and(b)the influence of participant characteristics and intervention protocols on these effects.Methods Searches were conducted in PubMed,Web of Science(Core Collection),Cochrane Library(Embase,ClinicalTrials.gov,and International Clinical Trials Registry Platform),and Chinese National Knowledge Infrastructure on April 2,with updates on October 17,2024.Pooled effects for each outcome were summarized using Hedge's g(g)through meta-analysis-based random effects models,and subgroup and regression analyses were used to explore moderators.Results A total of 24 studies with 621 participants were included.IT combined with BFR(IT+BFR)significantly improved maximal oxygen uptake(VO2_(max))(g=0.63,I^(2)=63%),mean power during the Wingate 30-s test(g=0.70,I^(2)=47%),muscle strength(g=0.88,I^(2)=64%),muscle endurance(g=0.43,I^(2)=0%),time to fatigue(g=1.26,I^(2)=86%),and maximal aerobic speed(g=0.74,I^(2)=0%)compared to IT alone.Subgroup analysis indicated that participant characteristics including training status,IT intensity,and IT modes significantly moderated VO2_(max)(subgroup differences:p<0.05).Specifically,IT+BFR showed significantly superior improvements in VO2_(max)compared to IT alone in trained individuals(g=0.76)at supra-maximal intensity(g=1.29)and moderate intensity(g=1.08)as well as in walking(g=1.64)and running(g=0.63)modes.Meta-regression analysis showed cuff width(β=0.14)was significantly associated with VO2_(max)change,identifying 8.23 cm as the minimum threshold required for significant improvement.Subgroup analyses regarding muscle strength did not reveal any significant moderators.Conclusion IT+BFR enhances physiological adaptations and optimizes aspects of endurance performance,with moderators including training status,IT protocol(intensity,mode,and type),and cuff width.This intervention addresses various IT-related challenges and provides tailored protocols and benefits for diverse populations.
基金Teaching and Research Project of Anhui Urban Management Vocational College(Project No.:2024kfkc001)。
文摘This paper reports a case of cerebral stem infarction with quadriplegia and complete dependence on daily life.The course of the disease lasted more than 7 months.Frenchay's improved articulation Disorder Assessment Form has been assessed as severe articulation disorder.The patient has significantly improved his speech function and quality of life after systematic head control training,respiratory function training,articulation motor training,and articulation training.In the course of treatment,emphasis was placed on head postural control training and respiratory function training,and emphasis was placed on the strength and coordination training of articulatory organs,and the results were remarkable.After the patient was discharged from the hospital,the follow-up of basic daily life communication was not limited.
基金supported by the National Natural Science Foundation of China(42374134,42304125,U20B6005)the Science and Technology Commission of Shanghai Municipality(23JC1400502)the Fundamental Research Funds for the Central Universities.
文摘Blended acquisition offers efficiency improvements over conventional seismic data acquisition, at the cost of introducing blending noise effects. Besides, seismic data often suffers from irregularly missing shots caused by artificial or natural effects during blended acquisition. Therefore, blending noise attenuation and missing shots reconstruction are essential for providing high-quality seismic data for further seismic processing and interpretation. The iterative shrinkage thresholding algorithm can help obtain deblended data based on sparsity assumptions of complete unblended data, and it characterizes seismic data linearly. Supervised learning algorithms can effectively capture the nonlinear relationship between incomplete pseudo-deblended data and complete unblended data. However, the dependence on complete unblended labels limits their practicality in field applications. Consequently, a self-supervised algorithm is presented for simultaneous deblending and interpolation of incomplete blended data, which minimizes the difference between simulated and observed incomplete pseudo-deblended data. The used blind-trace U-Net (BTU-Net) prevents identity mapping during complete unblended data estimation. Furthermore, a multistep process with blending noise simulation-subtraction and missing traces reconstruction-insertion is used in each step to improve the deblending and interpolation performance. Experiments with synthetic and field incomplete blended data demonstrate the effectiveness of the multistep self-supervised BTU-Net algorithm.
基金supported by a grant from University of Social Welfare and Rehabilitation Sciences for the research expenses.
文摘Background Schizophrenia is characterised by pervasive cognitive deficits that significantly impair daily functioning and quality of life.Pharmacological treatments have limited efficacy in addressing these deficits,highlighting the need for adjunctive interventions like computerised cognitive training(CCT).Aims This study aimed to evaluate the effects of a 30-session CCT programme on mental well-being and cognitive performance in individuals with schizophrenia.Additionally,it assessed the usability and acceptability of CCT in this population.Methods A double-blind,randomised clinical trial was conducted with 54 participants assigned to intervention and control groups.Cognitive and mental health outcomes were assessed using validated tools such as the Depression Anxiety Stress Scale 21,the Warwick-Edinburgh Mental Wellbeing Scale and the Cambridge Neuropsychological Test Automated Battery.Usability was measured with the System Usability Scale(SUS).Assessments were conducted at baseline,post-intervention and 3 months post-follow-up.Results The CCT intervention significantly improved mental well-being,reduced stress and enhanced working memory(paired associate learning,spatial working memory and spatial span)compared with controls.However,no significant effects were observed for anxiety,depression or executive function.Usability scores were high(SUS=83.51),and compliance rates were strong(92.7%),indicating favourable participant engagement.Conclusion CCT demonstrated potential as an adjunctive treatment for schizophrenia,with significant improvements in targeted cognitive and mental health domains.The high usability and compliance rates support its feasibility for broader implementation.Further research is needed to optimise protocols and explore long-term benefits.CCT offers a promising approach to addressing mental health and cognitive challenges in schizophrenia,particularly for stress and working memory.Its usability and acceptability suggest it could be seamlessly integrated into clinical practice.
基金the Talent Fund of Beijing Jiaotong University(Grant No.2024XKRC055).
文摘In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.
基金supported by a research grant from the Medical Research Program of Chongqing University Three Gorges Hospital(No.2022YJKYXM-017).
文摘Objective:The objective of this study is to develop and evaluate a structured clinical rotation-based emergency response training program to enhance nurses’emergency competence,theoretical knowledge,and technical skills.Methods:A comprehensive emergency training program was developed,and a randomized controlled trial was implemented from June 2022 to May 2023 at a tertiary general hospital in Chongqing,China.The study involved 214 nurses,with 106 participants in the intervention group receiving a 3-month innovative emergency response competence training and 108 in the control group undergoing conventional training.Postintervention assessments evaluated emergency response capabilities using the Emergency Response Ability Assessment Scale for Nurses in Public Health Emergencies,theoretical knowledge through a self-designed comprehensive theoretical assessment instrument,technical skills using a standardized skill assessment form,and training satisfaction through two distinct feedback instruments.Results:The emergency response ability scores were significantly higher in the intervention group compared to controls(3.99±0.18 vs.2.53±0.25,P<0.05).Theoretical assessment scores showed marked improvement in the intervention group versus the control group(85.31±4.40 vs.52.45±6.19,P<0.05).Technical skill performance was significantly better in the intervention group than that in controls(94.47±1.64 vs.86.39±2.36,P<0.05).Training satisfaction was higher among intervention group nurses compared to controls(4.53±0.23 vs.4.00±0.38,P<0.05),with nursing managers also reporting greater satisfaction with the intervention program versus conventional training(4.57±0.49 vs.3.92±0.79,P<0.05).Conclusion:The clinical rotation-based structured emergency response training program effectively enhances nurses’emergency competencies,theoretical knowledge,and technical skills.These findings provide both theoretical foundations and practical guidelines for developing emergency response and specialized nursing competence training programs.
文摘Endoscopic ultrasound(EUS)is an indispensable tool for the diagnosis and management of various diseases,particularly biliopancreatic disorders,as it provides detailed visualization of the gastrointestinal tract and surrounding structures.As the demand for diagnostic and interventional EUS procedures increases,ensuring high-quality training for endoscopists is essential to improve patient outcomes.This mini-review provides an overview of the current state of EUS training and emphasizes the importance of a structured approach that integrates theoretical knowledge and hands-on experience.We discuss different training methods,focusing on the main courses available worldwide,and highlight their advantages and limitations.In addition,we examine the challenges of training for diagnostic and interventional EUS,such as limited access to training centers and the need for personalized feedback.Overall,improving EUS training programs is essential to enhance physician skills and ensure this advanced technique is used safely and efficiently in clinical practice.
基金supported by the National Natural Science Foundation of China(62276092,62303167)the Postdoctoral Fellowship Program(Grade C)of China Postdoctoral Science Foundation(GZC20230707)+3 种基金the Key Science and Technology Program of Henan Province,China(242102211051,242102211042,212102310084)Key Scientiffc Research Projects of Colleges and Universities in Henan Province,China(25A520009)the China Postdoctoral Science Foundation(2024M760808)the Henan Province medical science and technology research plan joint construction project(LHGJ2024069).
文摘Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely used in pulmonary disease diagnosis,such as pneumonia and tuberculosis.However,traditional feature fusion methods often suffer from feature disparity,information loss,redundancy,and increased complexity,hindering the further extension of DL algorithms.To solve this problem,we propose a Graph-Convolution Fusion Network with Self-Supervised Feature Alignment(Self-FAGCFN)to address the limitations of traditional feature fusion methods in deep learning-based medical image classification for respiratory diseases such as pneumonia and tuberculosis.The network integrates Convolutional Neural Networks(CNNs)for robust feature extraction from two-dimensional grid structures and Graph Convolutional Networks(GCNs)within a Graph Neural Network branch to capture features based on graph structure,focusing on significant node representations.Additionally,an Attention-Embedding Ensemble Block is included to capture critical features from GCN outputs.To ensure effective feature alignment between pre-and post-fusion stages,we introduce a feature alignment loss that minimizes disparities.Moreover,to address the limitations of proposed methods,such as inappropriate centroid discrepancies during feature alignment and class imbalance in the dataset,we develop a Feature-Centroid Fusion(FCF)strategy and a Multi-Level Feature-Centroid Update(MLFCU)algorithm,respectively.Extensive experiments on public datasets LungVision and Chest-Xray demonstrate that the Self-FAGCFN model significantly outperforms existing methods in diagnosing pneumonia and tuberculosis,highlighting its potential for practical medical applications.
基金supported by Brazilian agencies CAPES(Finance Code 001)CNPq through PQ productivity scholarship.
文摘Here we compare the efficacy of anti-obesity drugs alone or combined with exercise training on body weight and exercise capacity of obese patients.Randomized clinical trials that assessed the impact of any anti-obesity drug alone or combined with exercise training on body weight,body fat,fat-free mass and cardiorespiratory fitness in obese patients were retrieved from Pubmed and EMBASE up to May 2024.Risk of bias assessment was performed with RoB 2.0,and the GRADE approach assessed the certainty of evidence(CoE)of each main outcome.We included four publications summing up 202 patients.Two publications used orlistat as an anti-obesity drug treatment,while the other two adopted GLP-1 receptor agonist(liraglutide or tirzepatide)as a pharmacotherapy for weight management.Orlistat combined with exercise was superior to change body weight(mean difference(MD):−2.27 kg;95%CI:−2.86 to−1.69;CoE:very low),fat mass(MD:−2.89;95%CI:−3.87 to−1.91;CoE:very low),fat-free mass(MD:0.56;95%CI:0.40–0.72;CoE:very low),and VO_(2)Peak(MD:2.64;95%CI:2.52–2.76;CoE:very low).GLP-1 receptor agonist drugs combined with exercise had a great effect on body weight(MD:−3.96 kg;95%CI:−5.07 to−2.85;CoE:low),fat mass(MD:−1.76;95%CI:−2.24 to−1.27;CoE:low),fat-free mass(MD:0.50;95%CI:−0.98 to 1.98;CoE:very low)and VO_(2)Peak(MD:2.47;95%CI:1.31–3.63;CoE:very low).The results reported here suggest that exercise training remains an important approach in weight management when combined with pharmacological treatment.
基金supported in part by the National Natural Science Foundation of China under Grants 62071345。
文摘Self-supervised monocular depth estimation has emerged as a major research focus in recent years,primarily due to the elimination of ground-truth depth dependence.However,the prevailing architectures in this domain suffer from inherent limitations:existing pose network branches infer camera ego-motion exclusively under static-scene and Lambertian-surface assumptions.These assumptions are often violated in real-world scenarios due to dynamic objects,non-Lambertian reflectance,and unstructured background elements,leading to pervasive artifacts such as depth discontinuities(“holes”),structural collapse,and ambiguous reconstruction.To address these challenges,we propose a novel framework that integrates scene dynamic pose estimation into the conventional self-supervised depth network,enhancing its ability to model complex scene dynamics.Our contributions are threefold:(1)a pixel-wise dynamic pose estimation module that jointly resolves the pose transformations of moving objects and localized scene perturbations;(2)a physically-informed loss function that couples dynamic pose and depth predictions,designed to mitigate depth errors arising from high-speed distant objects and geometrically inconsistent motion profiles;(3)an efficient SE(3)transformation parameterization that streamlines network complexity and temporal pre-processing.Extensive experiments on the KITTI and NYU-V2 benchmarks show that our framework achieves state-of-the-art performance in both quantitative metrics and qualitative visual fidelity,significantly improving the robustness and generalization of monocular depth estimation under dynamic conditions.
文摘Digital economy has become a new driving force for China’s economic growth,continuously boosting economic development and rapidly integrating into various fields of China’s economy and society.The advent of the digital economy era has reshaped the development pattern of the financial industry.The rapid development of financial technology has promoted the transformation of financial formats and put forward higher requirements for financial talent training in the new era.Digital finance is not only a key part of the transformation and upgrading of China’s financial industry but also an integral component of China’s modern financial ecosystem.Against the backdrop of the digital economy,cultivating digital finance talents in vocational colleges is of great significance to China’s economic development.This paper analyzes the predicaments faced in digital finance talent training,explores in depth the reform of digital finance talent training modes,and proposes policy suggestions for the digital finance talent training system based on the development characteristics of digital finance.
文摘High dropout rates in short-term job skills training programs hinder workforce development.This study applies machine learning to predict program completion while addressing class imbalance challenges.A dataset of6548 records with 24 demographic,educational,program-specific,and employment-related features was analyzed.Data preprocessing involved cleaning,encoding categorical variables,and balancing the dataset using the Synthetic Minority Oversampling Technique(SMOTE),as only 15.9% of participants were dropouts.six machine learning models-Logistic Regression,Random Forest,SupportVector Machine,K-Nearest Neighbors,Naive Bayes,and XGBoost-were evaluated on both balanced and unbalanced datasets using an 80-20 train-test split.Performance was assessed using Accuracy,Precision,Recall,F1-score,and ROC-AUC.XGBoost achieved the highest performance on the balanced dataset,with an F1-score of 0.9200 and aROC-AUC of0.9684,followed by Random Forest.These findings highlight the potential of machine learning for early identification of dropout trainees,aiding in retention strategies for workforce training.The results support the integration of predictive analytics to optimize intervention efforts in short-term training programs.
